Inhibition of Acute-, Latent-, and Chronic-Phase Human Immunodeficiency Virus Type 1 (HIV-1) Replication by a Bistriazoloacridone Analog That Selectively Inhibits HIV-1 Transcription

Abstract

Nanomolar concentrations of temacrazine (1,4-bis[3-(6-oxo-6H-v-triazolo[4,5,1-de]acridin-5-yl)amino-propyl]piperazine) were discovered to inhibit acute human immunodeficiency virus type 1 (HIV-1) infections and suppress the production of virus from chronically and latently infected cells containing integrated proviral DNA. This bistriazoloacridone derivative exerted its mechanism of antiviral action through selective inhibition of HIV-1 transcription during the postintegrative phase of virus replication. Mechanistic studies revealed that temacrazine blocked HIV-1 RNA formation without interference with the transcription of cellular genes or with events associated with the HIV-1 Tat and Rev regulatory proteins. Although temacrazine inhibited the in vitro 3′ processing and strand transfer activities of HIV-1 integrase, with a 50% inhibitory concentration of approximately 50 nM, no evidence of an inhibitory effect on the intracellular integration of proviral DNA into the cellular genome during the early phase of infection could be detected. Furthermore, temacrazine did not interfere with virus attachment or fusion to host cells or the enzymatic activities of HIV-1 reverse transcriptase or protease, and the compound was not directly virucidal. Demonstration of in vivo anti-HIV-1 activity by temacrazine identifies bistriazoloacridones as a new class of pharmaceuticals that selectively blocks HIV-1 transcription.

Current strategies for the therapeutic treatment of human immunodeficiency virus (HIV) infection/and AIDS are based upon the use of combination therapy with one or more inhibitors of the HIV type 1 (HIV-1) reverse transcriptase and protease enzymes. Even though combination therapies have achieved suppression of the viral loads in the sera of many patients, issues of multidrug resistance, compliance, and contraindications with these drug regimens continue to raise concerns (12). Moreover, combination therapies have revealed a need to address the issue of virus production from the long-lived populations of infected cells (11, 29). The exact nature of this infected-cell reservoir is unclear, but it may represent latently or chronically infected cells as well as persistent pockets of acutely infected cells. Thus, effective management of HIV-1 disease requires either long-term maintenance of patients on expensive and potentially toxic multidrug regimens or the development of new antiviral agents which target a broad spectrum of viral replication scenarios.

Antiviral agents directed toward viral targets that function after HIV-1 integration (during the postintegrative or late phase of viral replication) would be expected to inhibit virus production by acutely infected cells as well as from latently and chronically infected cells. Inhibitors of protease and the nucleocapsid protein (NCp7) zinc fingers have already been defined as inhibitors of very late events (posttranscriptional and translational) in the postintegrative phase of virus replication (24, 33). However, very few drugs target early events in the postintegrative phase, which includes transcription regulation by the HIV-1 regulatory proteins (37). Bisimidazoacridones and the related bistriazoloacridones, which were found to be potent and selective antitumor agents (8, 18), are thought to exert their cytotoxic actions by targeting a component of transcriptional regulation (28). One analog in the triazolozoacridones series, 1,4-bis[3-(6-oxo-6H-v-triazolo[4,5,1-de]acridin-5-yl)amino-propyl]piperazine (NSC 687025), now referred to as temacrazine (see Fig. 1A), was found to inhibit HIV-1 replication in cells acutely, chronically, and latently infected with HIV-1 when it was used at nanomolar concentrations. Mechanistic and cell-based assays defined temacrazine as a selective inhibitor of HIV-1 transcription.

FIG. 1.

(A) Structure of temacrazine. (B) Inhibition of HIV-1 expression in latently and chronically infected cell lines. U1 or ACH-2 cells were simultaneously treated with various concentrations of temacrazine and 5 ng of recombinant TNF-α per ml. Cultures were continued for 72 h, and cell-free supernatant was harvested for determination of virion-associated p24 levels. Alternatively, chronically infected H9/HTLV-III_B NIH 1983 cells were incubated for 5 days with the indicated concentrations of temacrazine, after which time the cell-free supernatants were harvested for determination of virion-associated reverse transcriptase activity. Viability was determined in all cases by the XTT dye reduction assay. Open symbols, cellular viability; closed symbols, virion-associated p24 or reverse transcriptase activity. The cell types used were U1 (squares), ACH-2 (circles), and H9/HTLV-III_B NIH 1983 (triangles). (C) Persistence of antiviral effects of temacrazine. U1 cells were cultured for 30 min with various doses of temacrazine and 5 ng of recombinant TNF-α per ml. The cultures were then washed five times with RPMI 1640 supplemented with 10% fetal calf serum and were continued in the absence or presence of additional TNF-α for 10 days. At day 10 the cell-free supernatants were collected for determination of virion-associated p24 levels, and viability was determined by XTT dye reduction assay. ○, viability; •, p24 levels with no additional TNF-α; ▪, p24 levels after the readdition of TNF-α. (D) Temacrazine inhibition requires the presence of viral elements. HeLa–CD4–LTR–β-gal cells were either pretreated with 100 nM temacrazine 2 h prior to virus adsorption (with washing to remove residual temacrazine) or exposed to temacrazine after 2 h of virus adsorption. Then the cultures were incubated for 48 h, fixed with 2% formaldehyde–2% glutaraldehyde (5 min), washed, and stained with 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside substrate for 50 min at 37°C. Blue cells (each representing a single infectious virion) were counted, and the number of infectious units per milliliter of sample was determined and expressed as a percentage of that for the control. Control cultures were treated during virus adsorption with 25 μM 2,2′-dithiobisbenzamide derivative, DIBA-1 (33), an NCp7 inhibitor.

MATERIALS AND METHODS

Synthesis of temacrazine.

The synthesis of temacrazine followed the procedures described earlier for the other members of the series (8). Briefly, the 5-chlorotriazoloacridone was condensed with 1,4-bis(3-aminopropyl)piperazine in dimethyl sulfoxide (DMSO) at 100°C in the presence of diisopropylethylamine. After precipitation with water, the crude product was purified by dissolving it in a 1% solution of methanesulfonic acid in water, and the solution was filtered to remove the undissolved unreacted material. The filtrate was made alkaline, and the resulting precipitate was crystallized (twice) from dimethylacetamide to yield the free base of temacrazine (60% yield, mp 242 to 245°C). Purity (+99%) was confirmed by high-pressure liquid chromatography (HPLC). Results of elemental analysis (for C₃₆H₃₄N₁₀O₂ · 0.5 H₂O) for C, H, and N were within ±0.4%. ¹H nuclear magnetic resonance chemical shifts of protons indicated in bold (CDCl₃): 9.43 (t, 2H, J = 5.5, NHCH₂), 8.55 (dd, 2H, C10-H), 8.52 (dd, 2H, C7-H), 8.16 (d, 2H, J = 9.2, C3-H), 7.88 (m, 2H, C9-H), 7.61 (m, 2H, C8-H), 7.0 (d, 2H, J = 9.2, C4-H), 3.61 (m, 4H, NHCH₂OCH₂), 2.55 (m, 12H, piperazine H and CH₂CH₂-piperazine), 1.99 (m, 4H, CH₂CH₂CH₂). Additionally, the structure was verified by single crystal X-ray structure analysis (to be published elsewhere).

The water-soluble dimethanesulfonate derivative was prepared from the free base by dissolving the free base in chloroform and precipitating the salt with methanesulfonic acid (mp 240 to 246°C decomp.). Results of elemental analysis for C, H, N, and S were within ±0.4% (for C₃₆H₃₄N₁₀O₂ · 2 CH₃SO₃H · 0.5 H₂O). ¹H nuclear magnetic resonance (DMSO-d₆): 9.42 (br, 2H, NHCH₂), 8.50 (dd, 2H, C10-H), 8.40 (dd, 2H, C7-H), 8.37 (d, 2H, J = 9.2, C3-H), 8.01 (m, 2H, C9-H), 7.69 (m, 2H, C8-H), 7.23 (d, 2H, J = 9.2, C4-H). The signals of the aliphatic protons were overlapped with the broad signals of water and DMSO and were not suitable for assignments (mass spectrum, m/e = 639).

Virus replication inhibition assays.

The effect of temacrazine on HIV-1 replication was determined with various cell types and HIV-1 isolates by either the 2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-5-[(phenylamino)carbonyl]-2H-tetra zolium hydroxide (XTT) cytoprotection assay as described previously (31, 32) or determination of virion-associated p24 levels in cell-free culture supernatants by an antigen-capture enzyme-linked immunosorbent assay (the assay kits were purchased from the AIDS Vaccine Program, National Cancer Institute [NCI]-Frederick Cancer Research and Development Center [FCRDC], Frederick, Md.). Phytohemagglutinin-stimulated human peripheral blood lymphocytes and monocyte/macrophages were prepared and used in antiviral assays as described previously (31). Temacrazine (NSC 687025), DIBA-1 (NSC 654077), and various control agents were obtained from the NCI chemical repository. The U1, ACH-2, CEM-SS, HeLa–CD4–LTR–β-gal, and H9/HTLV-III_B NIH 1983 cell lines were obtained from the AIDS Research and Reference Reagent Program (National Institute of Allergy and Infectious Diseases, Bethesda, Md.).

HIV-1 expression from TNF-α-inducible and chronically infected cell lines.

Compounds which interact with targets in the late or postintegrative phase of HIV-1 replication can best be identified with infected cells lines constitutively producing virus or latently infected cell lines in which virus expression from integrated provirus is induced by cytokines (10, 16, 30). U1 or ACH-2 cells (5 × 10⁴ cells/ml) were induced with 5 ng of tumor necrosis factor alpha (TNF-α) per ml for 24 h, followed by the addition of temacrazine, and were cultured for 72 h. The amount of virion-associated p24 was then determined in cell-free supernatants. Chronically infected H9/HTLV-III_B NIH 1983 cells (5 × 10⁴ cells/ml) were cultured in 25-cm² flasks for 5 days with various concentrations of temacrazine. Cell-free supernatants were harvested, and the virion-associated reverse transcriptase activity was determined. In all experiments cellular viability was determined by the XTT dye reduction assay.

Generation and use of chronically infected temacrazine-resistant, CEM-SS cells.

Temacrazine-resistant HIV-1_IIIB was generated by serial passage of virus in escalating doses of temacrazine. We are presently cloning and sequencing virus generated by the resistance generation protocol and have yet to identify mutations which account for the loss of antiviral activity reported here. The genetic analysis of the resistant virus along with reconstructed proviral clones will be published separately. Supernatants containing either wild-type virus or virus expressing more than a 1,000-fold resistance to temacrazine were used to infect naive CEM-SS cells. Cultures were continued until evidence of the presence of chronically infected cells was verified by the persistence of reverse transcriptase activity in the culture supernatants (approximately 2 to 3 weeks). These populations of chronically infected cells were cultured for 48 h with various concentrations of temacrazine, at which time the supernatants were harvested for determination of reverse transcriptase activity. Cell viability was determined by the XTT dye reduction assay.

Determination of HIV-1 protein and mRNA expression in U1 cells.

HIV-1 protein expression in the presence of temacrazine was evaluated by Western blotting as described previously (38). Briefly, a 50-μg protein equivalent of TNF-α-induced U1 cells in 5% β-mercaptoethanol was resolved on 4 to 20% polyacrylamide gels with sodium dodecyl sulfate (SDS) in Tris-glycine. After electroblotting onto polyvinylidene difluoride membranes, HIV-1 proteins were detected with a mixture of anti-HIV-1 NCp7 and anti-p24 antisera (kind gift of L. E. Henderson, NCI-FCRDC) and by chemiluminescence.

RNA was isolated from TNF-α-stimulated, temacrazine-treated U1 cells by the RNAzol method (Tel-Test Inc., Friendswood, Tex.) with a final lysate concentration of 10⁶ U1 cells in 1 ml of RNAzol. Ten micrograms of total RNA was electrophoresed in 0.75% agarose–2.2 M formaldehyde gels in 1× morpholinepropanesulfonic acid (MOPS) and was osmotically transferred to nylon membranes and cross-linked with UV light. Riboprobe-based detection of HIV-1-specific mRNA was carried out as specified in the in vitro transcription kit (Promega, Madison Wis.) by using a long terminal repeat (LTR)-specific riboprobe consisting of the U3-R-U5 regions of the pNL4-3 molecular clone. β-Actin mRNA was detected with ³²P-labeled, randomly primed pd(N)₆ (Gibco, Life Technologies, Gaithersburg, Md.) oligonucleotides from cloned cDNA. Hybridization was carried out overnight at 42°C, followed by washing for 10 min at ambient temperature in 2× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate)–0.1% SDS, 20 min at ambient temperature in 0.1% SSC–0.1% SDS, and 30 min at 55°C in 0.1× SSC–0.1% SDS. HIV-1-specific RNA expression and β-actin expression were documented by autoradiography.

Virus attachment and enzymatic assays.

The binding of HIV-1_RF to CEM-SS cells was measured by a p24-based assay (32). The effects of temacrazine on the in vitro activity of purified HIV-1 p66/51 reverse transcriptase (kind gift of S. Hughes, ABL-Basic Research Program, NCI-FCRDC), was determined by measurement of the level of incorporation of [³²P]TTP onto the poly(rA)-oligo(dT) (rAdT) or [³²P]GTP onto the poly(rC)-oligo(dG) (rCdG) homopolymer template-primer systems (31). HIV-1 protease activity was quantitated by a reversed-phase HPLC assay with the artificial Ala-Ser-Glu-Asn-Tyr-Pro-Ile-Val-Glu-amide HIV-1 protease substrate as described previously (31, 33). Ejection of zinc from recombinant NCp7 was determined by a fluorescence-based assay as described previously (31, 33, 34). The in vitro effect of temacrazine on HIV-1 integrase activity was determined by minor modifications of the procedure described by Bushman and Craigie (5) and Rice et al. (31). Purified recombinant HIV-1 integrase was a kind gift of S. Hughes (ABL-Basic Research Program, NCI-FCRDC). NSC 651016 (9) served as a known inhibitor for virus entry into cells, zidovudine triphosphate (Sierra BioResearch, Tuscon, Ariz.) and UC38 (NSC 629243) (3) inhibit HIV-1 reverse transcriptase with the rAdT and rCdG template-primers, respectively, KNI-272 (NSC 651714) (23) inhibits HIV-1 protease, ISIS 5320 (NSC 665353) (4) inhibits HIV-1 integrase, and NSC 624151 (dithiane) (31) causes the ejection of zinc from the HIV-1 NCp7 protein.

CAT assays.

LTR-mediated induction of chloroamphenicol acetyltransferase (CAT) in BF-24 cells (15) was measured by a fluorescence thin-layer chromatography assay (Flash-CAT; Stratagene, La Jolla, Calif.) 24 h after stimulation with either 5 ng of recombinant TNF-α per ml or 10 μg of recombinant Tat protein (Intracel, Cambridge, Mass.) alone or in combination with temacrazine. Seventy-five micrograms of cellular protein was analyzed for CAT activity as directed by the manufacturer of the assay kit, and activity was documented photographically with long-wavelength UV light. Positive and negative controls consisted of the protein-free reaction mixture with or without 5 U of purified recombinant CAT enzyme, respectively.

Cotransfection of Tat and Rev expression vectors with LTR reporter plasmids.

Cotransfections were carried out with 10⁶ cells (293 cells) with either 3 μg of CAT reporter plasmid and 1 μg of Tat, Rev, or empty expression plasmid by the calcium phosphate method. Temacrazine at various doses (1 to 1,000 nM) was added immediately after transfection, and cultures were harvested for CAT determination at 40 h posttransfection. CAT activity was determined by the solvent partition method (26). The plasmids used for cotransfections were pUXcat-RSV (in which constitutive CAT expression is controlled by the Rous sarcoma virus LTR; positive control), pRSPA (an empty expression vector; DNA control), pUXcat-HIV (in which the CAT gene is under the control of the HIV-1 LTR; Tat responsive), pDM-128 (in which the CAT gene is located in the intron of an HIV-1 subgenomic fragment containing the Rev responsive element (RRE); CAT gene expression is dependent on the Rev-RRE interaction), pRSA-HTat (which expresses the HIV-1 Tat protein), and pRS-HRev (which expresses the HIV-1 Rev protein) (7, 21).

Intracellular integration assay.

Integrated HIV-1 proviral DNA was identified by a two-step PCR protocol which specifically amplifies integrated provirus (to be described in detail elsewhere). To accomplish this we used the findings of Stevens and Griffith (36) that the majority of proviral clones examined have genomic Alu sequences within 2 kb of the 3′ LTR and extended PCR techniques to produce biotin-selectable PCR products containing the LTR-genomic junction as templates for the indirect identification of integrated provirus. CEM-SS cells (0.5 × 10⁶ cells) were infected with HIV-1_RF for 24 h, and DNA was isolated after proteinase K digestion by phenol-chloroform extraction. One microgram of total DNA was amplified for 30 cycles (94°C for 30 s, with a ramp time of 30 s to 60°C for 60 s and 72°C for 480 s, with a final cycle of 7 min of extension at 72°C) in PCR buffer containing 10 mM Tris-HCl (pH 8.8), 1.5 mM MgCl₂, and 75 mM KCl (Opti-prime buffer #6; Stratagene), 7.5 mM ammonium sulfate, 250 μM (each) deoxynucleoside triphosphate (Perkin-Elmer Corp., Norwalk, Conn.), 0.5 U of Taq DNA polymerase (Perkin-Elmer Corp.), 0.5 U of Taq extender (Stratagene), 36 pM sense primer (env [5′-CCA CCG CCT GAG AGA CTT 3′]; positions 8514 to 8532) and 180 pM antisense biotinylated primer (Alu [5′-biotin-TGG GAT TAC AGG CGT TGA G 3′] [25]). The products with random lengths (1 to 5 kb) representing the LTR-genomic junction were purified by strepavidin-biotin chromatography by magnetic bead separation techniques (M-280; Dynal, Lake Success, N.Y.). Magnetic beads (10 μg per 100 μl of the PCR mixture) were blocked by sequential washes in 0.1% bovine serum albumin (BSA)–phosphate-buffered saline (PBS) (2×), 1% BSA-PBS (1×), 0.1 mg of pd(N)₆ (Pharmacia Biotech, Piscataway, N.Y.) per ml, 0.1% BSA-PBS (1×), and 0.1% BSA-PBS (3×) prior to selection. The random-length biotin-labeled PCR products (representing integrated provirus) were separated from all other contaminating forms of genomic and cellular DNA by incubation with the blocked beads for 30 min at room temperature with agitation, followed by five washes in 1 M NaCl–Tris-EDTA (pH 8.0). The washed beads were then resuspended in H₂O, and a second PCR was carried out with standard DNA PCR conditions (30 cycles of 94°C for 30 s, with ramp times of 30 s to 60°C for 60 s and 72°C for 60 s, with a final cycle of 7 min of extension at 72°C) and with Gene-Amp reagents (Perkin-Elmer Corp.) and primers located in the nef region of HIV-1 (sense primer [5′-CAA GTA TTG GTG GAA TCT CC-3′], positions 8589 to 8610; antisense primer [5′-TTG CCA CCC ATT TTA CAG-3′], positions 8794 to 8777) for detection of integrated provirus. Monitoring for genomic and episomal HIV-1 DNA contamination was done by performing PCR for gag DNA (primers M661 and M667 [40]). The PCR products were separated by 2% agarose gel electrophoresis, stained with ethidium bromide for identification of specific PCR products, and documented photographically. All oligomers derived from HIV-1 DNA sequences were obtained from previously published HIV-1_RF sequences found in GenBank (accession numbers M17451 and M12508).

In addition to PCRs for the detection of integrated provirus, PCR of DNA was also performed under standard conditions at 24 h after infection to verify the completion of reverse transcription (40). Antiviral activity was also determined with an aliquot of cells removed prior to lysis to retrieve DNA. Cells (10⁴) were placed in 96-well plates, and the plates were cultured for 7 days, after which antiviral activity was measured by the XTT cytoprotection assay. In all experiments temacrazine had no effect on the expression of late reverse transcription products and antiviral activity was maintained (data not shown).

In vivo antiviral activity.

The murine hollow-fiber model for in vivo inhibition of HIV-1 replication was performed as described previously (20). Briefly, CEM-SS cells were acutely infected with HIV-1_IIIB at a low multiplicity of infection. The infected cells were immediately loaded into hollow polyvinylidene fluoride fibers (inner diameter, 1 mm; (Spectrum Medical Corp., Houston, Tex.) with a molecular size exclusion of ≥500,000 Da. Three fiber cultures were implanted subcutaneously and three fiber cultures were implanted intraperitoneally into each severe combined immunodeficient mouse (NCI Animal Production Facility, NCI-FCRDC). For each experiment, groups of mice received treatment with the compound vehicle (physiological saline or DMSO), temacrazine (in DMSO), or the positive control compound, dideoxycytosine (ddC) in saline by the intraperitoneal route. The treatments were administered on days 0 (day of fiber implant) through 6, with samples collected on day 7. Fibers were sampled for cell viability determinations and reverse transcriptase quantitation, and serum and peritoneal washes were evaluated for p24 antigen quantitation. The p24 antigen concentrations were determined with a commercially available enzyme-linked immunosorbent assay kit (Coulter Diagnostics, Hialeah, Fla.). Cell viability was determined by a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) dye conversion assay, which has a stable endpoint. The Student t test was performed to determine statistical significance.

RESULTS

Antiviral properties of temacrazine.

Temacrazine (Fig. 1A) was a potent inhibitor of virus production from TNF-α-induced latently infected U1 or ACH-2 cell lines or from chronically infected H9 cells (Fig. 1B). Fifty percent inhibition of virus production occurred in the range of 0.1 to 10 nM, while cytotoxicity was observed at concentrations approximately 1,000-fold higher, in the range of 1 to 10 μM (Fig. 1B). Temacrazine inhibition of HIV-1 expression from U1 and ACH-2 cells was equally effective for both cell lines when temacrazine was added 24 h prior to, simultaneously with, or 24 h after induction with TNF-α (Fig. 1B and data not shown). Moreover, pulsing of U1 cells with temacrazine for 30 min resulted in a prolonged reduction in virion-associated p24 production (as long as 10 days of culture) (Fig. 1C).

Temacrazine also inhibited acute infections with all strains of HIV-1 tested, including strains resistant to the reverse transcriptase inhibitors zidovudine, nevirapine, and didanosine (Table 1). However, temacrazine failed to inhibit replication by HIV-2 or the simian immunodeficiency virus, demonstrating that the antiviral action of the compound is highly specific for HIV-1. Uninfected HeLa–CD4–LTR–β-gal cells were pulsed with temacrazine 2 h prior to the addition of infectious HIV-1 or were exposed to compound following a 2-h preadsorption of virus to cells, and infections were monitored by looking for the formation of blue colonies (Fig. 1D). Pretreatment of cells with temacrazine did not inhibit virus expression. In contrast, application of the compound after virus adsorption resulted in >90% inhibition of infection, even though temacrazine was not directly virucidal. Thus, the antiviral effects of temacrazine are highly specific for HIV-1 and appear to require the presence of viral components.

TABLE 1.

Range of antiviral action of temacrazine

Virus group and cell typea	Virus strainb	Temacrazinec			EC50 (nM) of AZT
		EC50 (nM)	IC50 (nM)	TI50
Laboratory HIV-1 isolates and CEM-SS cells	RF	1.1	2,770	2,518	2
	IIIB	5.6	2,770	495	2.5
Clinical HIV-1 isolates and PBLs	WEJO (SI)	10.0	>1,000	>100	3
	ROJO (SI)	30.0	>1,000	>33	11
	SLKA (NSI)	10.0	>1,000	>100	4
Monocytotropic HIV-1 isolates and Mo/Mφ	Ba-L	19.8	>2,000	>100	20
	ADA	5.0	>10,000	>2,000	30
HIV-1 clade isolates and PBLs	Clade A	15.2	>1,500	>98	1
	Clade B	5.6	>1,500	>268	6
	Clade C	6.4	>1,500	>234	20
	Clade D	72.0	>1,500	>21	15
	Clade E	8.7	>1,500	>172	50
	Clade F	7.4	>1,500	>203	2
Drug-resistant HIV-1 isolates and PBLs	AZT-R	10.0	>1,000	>100	1100
	nevirapine-R	10.0	>1,000	>100	50
	ddI-R	10.0	>1,000	>100	1

^aAnti-HIV studies with lymphocyte-derived cell lines were performed by the XTT cytoprotection assay. Antiviral assays with phytohemagglutinin-stimulated human peripheral blood lymphocytes (PBLs) and monocyte/macrophage (Mo/Mφ) cultures were performed by measuring p24 levels in the supernatant. All determinations of antiviral activity were carried out with viral stock dilutions resulting in >80% reduction in cell viability (cell lines) or >50% multinucleated giant cell formation (monocytes/macrophages) at 5 or 12 days of culture, respectively. Inhibition of HIV-1 replication in all cases was dependent on the multiplicity of infection (ratio of infectious virus to host cell). 

^bSI and NSI, syncytium-inducing and non-syncytium-inducing variants of HIV-1, respectively; R, viral isolates resistant to the indicated compounds. 

^cEC₅₀, concentration which inhibits viral replication by 50%, IC₅₀, concentration which reduces cell viability by 50%; TI₅₀, EC₅₀/IC₅₀. 

Temacrazine selectively inhibits HIV-1 transcription.

Following viral integration, HIV-1 gene expression requires the transcription of proviral DNA to generate viral RNA and the subsequent synthesis and processing of viral proteins. Western blot analysis of viral protein production in TNF-α-induced U1 cells revealed that temacrazine caused blockage of the synthesis of Gag precursor polyproteins and mature viral proteins (Fig. 2A), which were mirrored by dramatic changes in HIV-1-specific RNA expression when total RNA was examined by Northern blotting (Fig. 2B). Induction of HIV-1 expression with TNF-α results in the upregulation of specific viral RNA synthesis, with increases in multiply spliced RNA (2.0 kb) expression and the appearance of singly spliced (4.3-kb) and unspliced (9.2-kb) RNA species (16) (Fig. 2B, lane 2). Treatment of induced U1 cells with 100 or 50 nM temacrazine completely suppressed the expression of singly spliced and unspliced viral RNA species, while the amount of multiply spliced RNA species was significantly reduced, but they were still present. Intermediate reductions in the levels of RNA were also seen with 10 and 5 nM temacrazine. These results were confirmed by mimic reverse transcription-PCR, in which all viral RNA species were suppressed by temacrazine (data not shown).

FIG. 2.

(A) Temacrazine prevents the synthesis of HIV-1 proteins. U1 cells were induced with 5 ng of TNF-α per ml for 24 h followed by the addition of temacrazine and the continuation of culturing for 72 h. The cells were then lysed, and proteins were processed for Western blotting and chemiluminescence detection as described in Materials and Methods. (B) Temacrazine inhibits HIV-1 RNA expression. Temacrazine-treated, TNF-α-induced U1 cells were lysed and total cellular RNA was isolated. HIV-1-specific RNA and β-actin mRNA were detected by Northern blotting as described in Materials and Methods. MW, molecular weight.

Inhibition of HIV-1 transcription was a selective event that was not accompanied by the global suppression of cellular transcription. Evaluation of cellular β-actin gene expression by Northern blotting revealed no effect of temacrazine on its transcription (Fig. 2B). The lack of effect on cellular housekeeping gene expression and general cellular transcription was further verified by reverse transcription-PCR for the detection of the single-copy gene, porphobilinogen deaminase (hydroxymethylbilane synthase) (35) (data not shown). Additionally, temacrazine did not inhibit [³H]leucine, [³H]thymidine, or [³H]uridine incorporation into CEM-SS cells after 24 h of exposure to temacrazine (concentrations required to suppress ³H incorporation by 50%, 620, 1,000, and >1,000 nM, respectively).

The specificity of action of temacrazine was further investigated by determining whether the compound would inhibit transcriptional activation of the HIV-1 LTR by TNF-α-inducible cellular transcriptional elements, such as NFκB. For this purpose, we used BF-24 cells, derived from the monocytic leukemia cell line THP-1, which stably expresses a CAT gene under the transcriptional control of the HIV-1 LTR (15). The addition of TNF-α resulted in enhanced CAT gene expression and enzymatic activity in BF-24 cells, but temacrazine at a high test concentration of 5 μM (1,000-fold greater than the 50% effective concentration) did not inhibit CAT expression (Table 2). Likewise, the high test concentration of temacrazine of 100 nM had no effect on either the quantity or the length of transcribed RNA when HeLa nuclear extracts were used for nuclear runoff transcription of a cytomegalovirus immediate-early promoter reporter gene construct (data not shown). The failure of temacrazine to inhibit cell-based and in vitro transcriptional systems in the absence of HIV-1 infection demonstrates that this compound’s action is specific for HIV-1 transcription.

TABLE 2.

Mechanism-of-action studies

Target	Evaluation procedure	I50 (nM)a
Tat	LTR-CAT reporter construct (BF-24 cells)	NI
	Cotransfection of Tat and LTR-CAT constructs	NI
Rev	Cotransfection of Rev and RRE-CAT constructs	NI
Virus attachment	p24-based assay	NI
Reverse transcriptase	rAdT and rCdG template-primer systems	NI
Protease	HPLC-based assay of substrate cleavage	NI
NCp7 Zn fingers	Fluorescence-based Trp37 assay	NI
Integrase	In vitro 3′ processing and strand transfer assay	10–100

^aI₅₀, concentration which results in 50% inhibition; NI, not inhibitory at a high test dose (100 μM). 

Evaluation of temacrazine against classical antiviral targets.

Because successful HIV-1 transcription and subsequent expression of viral transcripts require the participation of the virus-derived Tat and Rev regulatory proteins, we next determined if temacrazine affected the actions of either of these two proteins or their regulatory pathways. Temacrazine failed to inhibit transactivation of the LTR-CAT reporter construct in BF-24 cells by exogenous Tat or the interactions of Tat with the transactivation region (TAR) or Rev with the RRE in cotransfection assays with the appropriate CAT reporter constructs in 293 cells (7, 21) (Table 2). These findings suggest that temacrazine does not affect the Tat-TAR or Rev-RRE elements but that the mechanism of action of temacrazine may involve an unidentified transcriptional target. Temacrazine also did not inhibit other classical antiviral targets (reverse transcriptase, protease, the NCp7 zinc fingers, or virus attachment) and was not directly virucidal but inhibited the in vitro 3′ cleavage and strand transfer activities of HIV-1 integrase (Table 2).

Although the activity of temacrazine in postintegrative, late-phase model systems (Fig. 1B and C) suggested that the molecular target for temacrazine was independent of integrase (since integrase functions in the early phase of infection), we developed a qualitative PCR-based assay for the detection of the integrated provirus in acutely infected cells in order to assess the intracellular effects of temacrazine on integrase. However, no discernible alterations in proviral integration were observed in the presence of temacrazine (Fig. 3A and B). To confirm that temacrazine was acting on a postintegrative event, we selected for a temacrazine-resistant isolate (HIV-1_TR), used that virus isolate to develop a chronically infected cell line, and then tested the ability of temacrazine to inhibit virus production from those cells containing the integrated HIV-1_RT provirus. Figure 3C shows that temacrazine failed to inhibit viral replication in chronically infected cells expressing HIV-1_TR, while replication of wild-type virus (HIV-1_IIIB) in chronically infected cells (derived in parallel with the resistant virus-infected cells) was effectively suppressed. These data illustrate that temacrazine blocks a postintegrative transcriptional event by acting on an as yet undetermined molecular target.

FIG. 3.

In vitro inhibition of the HIV-1 integrase enzyme does not correlate with the intracellular effects of integrase. (A and B) Temacrazine does not inhibit HIV-1 proviral integration. Infected CEM-SS cells were infected with HIV-1_RF in the presence or absence of temacrazine at various doses. Following incubation for 24 h, total genomic DNA was isolated and was subjected to PCR for the detection of integrated provirus. (A) PCR with gag primers for the detection of genomic and episomal HIV-1 DNA contamination after selection and purification of the random-length products generated by the first PCR. (B) Detection of integrated provirus with primers internal to the first sense primer and the LTR-genomic DNA junction on the selected PCR product. (C) Effect of temacrazine on chronically infected CEM-SS cells expressing either temacrazine-resistant or wild-type HIV-1. Chronically infected CEM-SS cells expressing either wild-type or temacrazine-resistant HIV-1_IIIB were cultured for 48 h with the indicated concentrations of temacrazine. At 48 h cell-free supernatants were collected and virion-associated reverse transcriptase activity was determined. Viability was determined by the XTT dye reduction assay. Open symbols, cellular viability; closed symbols, reverse transcriptase activity; circles, chronically infected CEM-SS expressing wild type HIV-1_IIIB; squares, chronically infected CEM-SS cells expressing temacrazine-resistant HIV-1_IIIB.

The potential chemotherapeutic value of temacrazine was evaluated by testing the compound in a murine hollow-fiber in vivo model of HIV-1 replication. Hollow fibers containing HIV-1_IIIB-infected CEM-SS cells were implanted intraperitoneally and subcutaneously into severe combined immunodeficient mice, and then the mice were dosed intraperitoneally with temacrazine (Table 3). Temacrazine displayed in vivo anti-HIV-1 activity (reductions in p24 antigen levels) in both the peritoneal washes (from 2,451 to <500 pg of p24) and serum samples (from 3,716 to <2,000 pg of p24) at nontoxic concentrations.

TABLE 3.

In vivo activity of temacrazine in the hollow-fiber mouse model^a

Treatment	No. of animals	Concn (pg/ml) of p24 in the following sample:
		Peritoneal wash			Serum
		Mean	SD	Pb	Mean	SD	P
None	3	0	0		0	0
Saline every 8 h	5	2,185	675		2,694	717
DMSO every 8 h	5	2,451	868		3,716	1,287
ddC at 40 mg/kg every 8 h	3	7	12	0.002	87	21	0.0009
Temacrazine
8.5 mg/kg every 8 h	2	460	396	0.026	1,790	198	0.1
12.5 mg/kg every 12 h	3	190	97	0.004	1,863	601	0.06

^aHollow fibers containing HIV-1_IIIB-infected CEM-SS cells were implanted into nude mice, and the mice were dosed with temacrazine (total dosage, 25 mg/kg of body weight day for 7 days; total dose, 175 mg/kg). All compounds were administered intraperitoneally. 

^bStatistical comparison of treated groups with the DMSO vehicle control group by the Student t test. Boldface indicates a statistically significantly difference from the results for animals injected with DMSO alone. 

DISCUSSION

Our efforts to identify new antiviral agents targeting events immediately after proviral integration, with the aim of developing a broad-spectrum antiviral agent with activity against acutely, chronically, and latently HIV-1 infected cells, led to the discovery of temacrazine. This bistriazoloacridone is an inhibitor of HIV-1 replication in vitro when it is used at nanomolar concentrations and also exerts in vivo anti-HIV-1 activity. Additionally, temacrazine demonstrated a broad in vitro selectivity index (50% effective concentration/50% inhibitory concentration) of approximately 1,000. Temacrazine selectively inhibits HIV-1 transcription but had no measurable effects on cellular transcription at the doses that exhibited antiviral activity. Temacrazine-mediated inhibition was found to be independent of transcriptional pathways controlled by TNF-α, Tat-TAR, or Rev-RRE interactions. Thus, temacrazine affects a highly sensitive molecule involved in HIV-1 transcription that can be selectively targeted both in vitro and in vivo.

The retroviral replication cycle requires that proviral DNA integrate into the host cell genome, initiate the transcription of new virus, and lead ultimately to the production of infectious virions. The commonality of early transcription processes required for the production of virus from acutely, chronically, and latently infected cells suggests that agents which target this process would have an antiviral activity against a wide variety of replication-competent cell types and viral expression states. The search for agents that target transcription has led to the identification of inhibitors reported to affect the Tat protein and other chemotypes that inhibit HIV-1 transcription by unknown mechanisms (6, 17, 22). Among these, Baba et al. (2) recently reported a piperazinyloxoquinoline derivative (K12) with activity against acutely, chronically, and latently infected cells in the nanomolar range. Interestingly, their K12 compound is also active against HIV-2. This is in contrast to the high degree of specificity of temacrazine for all strains of HIV-1; it is not cross-reactive with HIV-2 and does not inhibit non-virus-based transcription. These observations suggest that temacrazine affects a highly specific viral target required for HIV-1 transcription. Recently, Andrews et al. (1) reported that the human T-cell lymphotropic virus type 1 uses an RNA polymerase complex with an α-amanitin inhibitory profile, suggestive of a hybrid of type II and type III RNA polymerase complexes. It is not known if the selectivity of temacrazine is due to an attack on a subset of unique accessory proteins in the RNA polymerase complex or if it reflects differential sensitivity conferred by the incorporation of viral control elements.

Although temacrazine did not interfere with the transactivation of a stably expressed LTR-CAT construct by recombinant Tat protein or interfere with Tat stimulation of an LTR construct in cotransfection experiments, the high levels of Tat protein (10 μg) required to transactivate BF-24 cells and the overexpression of Tat protein generated during the cotransfection experiments may account for the inability of temacrazine to inhibit the Tat-mediated transactivation in the experiments presented here. This is consistent with our finding that temacrazine is sensitive to the virus multiplicity of infection in assays for in vitro antiviral activity (data not shown). Interestingly, Baba et al. (2) also reported a similar effect for the K12 compound. Thus, our data do not completely rule out the possibility that temacrazine targets HIV-1 transcription by interfering either directly with Tat or indirectly by altering its association or the functional activities of the Tat-associated proteins in the transcriptional complex (13, 19, 39, 41).

As part of our studies with temacrazine, we determined its activity in a battery of cellular and molecular target-based assays. Temacrazine did not affect HIV-1 reverse transcriptase, protease, or NCp7, was not directly virucidal, and did not inhibit virus attachment or fusion. However, temacrazine was a potent inhibitor of HIV-1 integrase enzyme activity in the in vitro oligomer-based assay (5). This assay has classically identified a large number of compounds that inhibit integrase but that lack substantial antiviral activity in vitro (14, 27). Therefore, we sought to determine if temacrazine acted biologically by blocking the intracellular integration of proviral DNA in the cellular genome. Using a PCR-based assay to detect integrated provirus, we were unable to show any significant reduction in the level of proviral integration in the presence of temacrazine. Additionally, we generated a temacrazine-resistant HIV-1 isolate by serial passage and then used this virus and the corresponding wild-type (temacrazine-sensitive) virus to generate chronically infected cells. Temacrazine effectively inhibited wild-type virus in the chronically infected cells but failed to alter the expression of the resistant virus, showing that a major target for the action of temacrazine is present after HIV-1 proviral integration. Thus, we have found no evidence of antiviral action of temacrazine during the preintegrative phase of HIV-1 replication. On the other hand, we cannot rule out the possibility that temacrazine’s antiviral activity is totally independent of an action against HIV-1 integrase. Studies are under way to identify the molecular target for temacrazine-mediated transcriptional inhibition.

In summary, we have identified temacrazine as a highly selective inhibitor of HIV-1 RNA expression. Temacrazine inhibited in vitro HIV-1 replication in acutely, chronically, and latently infected cells and inhibited the in vivo replication of HIV-1 in an animal model. Moreover, we demonstrated that the inhibition of HIV-1 transcription did not correlate with an inhibitory action on Tat-TAR- or Rev-RRE-mediated events or general cellular transcription. Given these findings, temacrazine represents a novel class of antiviral agents for AIDS chemotherapy, as well as a pharmacologic tool for use in the development of new insights into transcriptional control during HIV-1 replication.